Theranostics 2014, Vol. 4, Issue 7 721

Ivyspring International Publisher Theranostics 2014; 4(7): 721-735. doi: 10.7150/thno.9052 Research Paper Luminescent Dual Sensors Reveal Extracellular pH-Gradients and Hypoxia on Chronic Wounds That Disrupt Epidermal Repair Stephan Schreml1, Robert J. Meier2, Michael Kirschbaum3, Su Chii Kong4, Sebastian Gehmert5, Oliver Felthaus6, Sarah Küchler7, Justin R. Sharpe8, Kerstin Wöltje3, Katharina T. Weiß1, Markus Albert1, Uwe Seidl1, Josef Schröder9, Christian Morsczeck6, Lukas Prantl5, Claus Duschl3, Stine F. Pedersen4, Martin Gosau6, Mark Berneburg1, Otto S. Wolfbeis2, Michael Landthaler1, Philipp Babilas1

1. Department of Dermatology, University Medical Center , Franz-Josef-Strauss-Allee 11, 93053 Regensburg, 2. Institute of Analytical Chemistry, Chemo- and Biosensors, University of Regensburg, Universitätsstraße 31, 93053 Regensburg, Germany 3. Fraunhofer Institute for Biomedical Engineering, Branch Potsdam, Am Mühlenberg 13, 14476 Potsdam, Germany 4. Department of Biology, University of Copenhagen, 13 Universitetsparken, DK-2100 Copenhagen, Denmark 5. Center of Plastic Surgery, Department of Trauma Surgery, University Medical Center Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Ger- many 6. Department of Maxillofacial Surgery, University Medical Center Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany 7. Department of Pharmacy, Freie Universität Berlin, Königin-Luise-Str. 2-4, 14195 Berlin, Germany 8. Blond McIndoe Research Foundation, Queen Victoria Hospital, Holtye Road, East Grinstead RH19 3DZ, England, UK 9. Center for Electron Microscopy at the Institute of Pathology, University Medical Center Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany

 Corresponding author: Stephan Schreml, M.D. Department of Dermatology, University Medical Center Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany. T: +49-(0)941-944-9614; F: +49-(0)941-944-9611; M: [email protected]

© Ivyspring International Publisher. This is an open-access article distributed under the terms of the Creative Commons License (http://creativecommons.org/ licenses/by-nc-nd/3.0/). Reproduction is permitted for personal, noncommercial use, provided that the article is in whole, unmodified, and properly cited.

Received: 2014.03.09; Accepted: 2014.04.04; Published: 2014.04.30

Abstract Wound repair is a quiescent mechanism to restore barriers in multicellular organisms upon injury. In chronic wounds, however, this program prematurely stalls. It is known that patterns of ex-

tracellular signals within the wound fluid are crucial to healing. Extracellular pH (pHe) is precisely regulated and potentially important in signaling within wounds due to its diverse cellular effects. Additionally, sufficient oxygenation is a prerequisite for cell proliferation and protein synthesis during tissue repair. It was, however, impossible to study these parameters in vivo due to the lack

of imaging tools. Here, we present luminescent biocompatible sensor foils for dual imaging of pHe and oxygenation in vivo. To visualize pHe and oxygen, we used time-domain dual lifetime refer- encing (tdDLR) and luminescence lifetime imaging (LLI), respectively. With these dual sensors, we

discovered centripetally increasing pHe-gradients on human chronic wound surfaces. In a thera- peutic approach, we identify pHe-gradients as pivotal governors of cell proliferation and migration, and show that these pHe-gradients disrupt epidermal barrier repair, thus wound closure. Parallel oxygen imaging also revealed marked hypoxia, albeit with no correlating oxygen partial pressure

(pO2)-gradient. This highlights the distinct role of pHe-gradients in perturbed healing. We also found that pHe-gradients on chronic wounds of humans are predominantly generated via cen- + + trifugally increasing pHe-regulatory Na /H -exchanger-1 (NHE1)-expression. We show that the modification of pHe on chronic wound surfaces poses a promising strategy to improve healing. The study has broad implications for cell science where spatial pHe-variations play key roles, e.g. in tumor growth. Furthermore, the novel dual sensors presented herein can be used to visualize pHe and oxygenation in various biomedical fields.

Key words: fluorescence imaging, wound healing, cell migration, gradient sensing, proton trans- porters

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 722

Introduction To address the questions raised above, we de- vised a 2D luminescent dual sensor for both pHe and In multicellular organisms, acute (i.e. physio- pO2 based on our previous sensor concepts [28-31]. logical) wound healing is a dormant mechanism to We subsequently used this sensor to visualize pO2- restore barrier integrity upon injury [1-3], and it and pHe-patterns in human chronic venous ulcers, comprises three major phases: (i) inflammation (e.g. which represent the vast majority of all chronic leukocyte recruitment, secretion of cyto- wounds. We then studied the impact of the detected kines/chemokines and growth factors), (ii) tissue pHe-values on keratinocytes (which are essential for formation (e.g. fibroblast and keratinocyte prolifera- restoring an epithelial cover) in terms of proliferation, tion/migration, neoangiogenesis), and (iii) remodel- viability and migration (2D/3D wound healing as- ing (e.g. by matrix metalloproteinases = MMPs) [1, says, microfluidic cell migration assays). Additional- 3-5]. In chronic wounds, however, this program ly, proton transporter expression was investigated in prematurely stalls. Chronic wounds affect approxi- biopsies from different regions in chronic wounds to mately 1-2% of the aging population in developed elucidate how the pHe-variations are generated. countries and thus pose an exceptional health and socio-economic issue [6]. About a third of the derma- Results tological health budget is spent on the treatment of chronic wounds, more than for all Luminescent dual sensors for imaging pO2 and dermato-onocological indications (melanoma and pHe in vivo other skin cancers) taken together [6, 7]. To visualize 2D pO2/pHe-distribution in vivo, we A plethora of factors may be the starting point devised a novel referenced luminescent dual sensor for chronic wound development [8, 9], yet little is (Fig. 1A) based on our previous sensor concepts [30, known about the precise mechanisms which prevent 31]. We used palladium(II)-meso-tetraphenyl- cells from restoring an intact epithelial cover. Under- tetrabenzoporphyrin in oxygen-permeable poly standing the spatial distribution of biological param- (styrene-co-acrylonitrile) microparticles (PdTPTBP- eters in healing is crucial to elucidate their impact on PSAN) as an oxygen probe [31, 32]. For pHe-imaging, the cells involved in wound repair. In recent years, pH-dependent fluorescein-isothiocyanate was cova- gradients of signaling molecules have been found to lently bound to aminoethylcellulose microparticles govern directional cell migration in acute wounds [8, (FITC-AC). The pH-independent reference dye 10-12]. In this context, the wound fluid plays a crucial ruthenium(II)-tris(4,7-diphenyl-1,10-phenanthroline) role [13-16] as spatial patterns of extracellular signals was incorporated in oxygen-impermeable polyacry- therein (e.g. cytokines, chemokines, and H2O2) are lonitrile microparticles (Ru(dpp)3-PAN) to prevent important in regulating wound repair [8, 10, 17]. Little luminescence quenching via oxygen [30]. Microparti- is known, however, about equivalent mechanisms cles were embedded in a biocompatible polyurethane possibly present in chronic wounds. hydrogel layer on transparent poly(vinylidene- Extracellular hydrogen ion concentration (pHe) chloride) (PVdC) foils to prevent dye/microparticle is precisely regulated and potentially important in exposure/uptake of cells, and to create a flexible 2D signaling within chronic wounds due to its diverse sensor [30, 31]. For details, see Materials and Meth- cellular effects [18-21]. Additionally, cells involved in ods. wound closure need suffient oxygen supply for up- Following photoexcitation, luminescence was regulated energy-consuming processes, such as pro- detected with a -coupled device (CCD)-camera. liferation, migration, and protein synthesis [19-22]. Signals were separated by optical filters (Fig. 1B,C), Both pO2 and pHe are essential metabolic parameters and by two referenced time-gated measurement which affect cell proliferation and migration [22, 23]. schemes (Fig. 1D,E). Luminescence lifetime imaging Our venture into this study was prompted by (LLI) [31] of integrated signal intensities of the fact that there is no satisfactory animal model for PdTPTBP-PSAN in two time gates during emission chronic wounds, and by the lack of tools for decay (Fig. 1D) yielded intrinsically referenced 2D-imaging of pHe and pO2 in vivo. For these reasons, pO2-values (Fig. 1F). Time-domain dual lifetime ref- the roles of pHe and pO2 in chronic healing of humans erencing (tdDLR) of integrated signal intensities dur- have hitherto remained poorly understood. The tools ing excitation (FITC-AC and Ru(dpp)3-PAN signals) available (glass electrode for pH, Clark electrode for and emission (only Ru(dpp)3-PAN signals) was used pO2) merely yield spot-measurements [24], and may for referenced spatial pHe-detection (Fig. 1E,G) [30]. not reveal possible spatial distribution patterns. In For details, see Materials and Methods. There was no contrast, luminescence imaging techniques are versa- relevant dye bleaching [30], cross-sensitivity or tem- tile tools to analyze spatial distribution patterns of perature-dependency of pO2/pHe-signals (Additional biological parameters in live organisms [10, 25-27].

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 723

File 1: Fig. S1). was kept thin (6 µm) in order to warrant fast response, The dual sensor described herein has a response in particular to pHe [28, 30, 31, 33]. It is also manda- time (99% signal intensity change) of ~1 min towards tory that sensor layers are well-conditioned (e.g. in both oxygen and pH as shown in previous works on Ringer´s solution, see Materials and Methods) prior sensors for pHe and pO2 [28, 30, 31]. Response time is to use. Oxygen diffuses quite rapidly through the mainly determined by the diffusion of oxygen and PSAN copolymer microparticles, which are entirely protons through the hydrogel layer, which therefore impermeable to protons.

Figure 1: Dual 2D luminescence imaging of pO2 and pHe in vivo. (A) Sensor foil scheme: Oxygen-dependent palladium(II)-meso-tetraphenyltetrabenzoporphyrin incorporated in oxygen-permeable poly(styrene-co-acrylonitrile) particles (Pd-TPTBP-PSAN), pH-sensitive fluorescein-isothiocyanate bound to aminocellulose particles (FITC-AC), and the pH-independent reference dye ruthenium(II)-tris(4,7-diphenyl-1,10-phenanthroline) in oxygen-impermeable polyacrylonitrile particles (Ru(dpp)3-PAN). Microparticles were embedded in a polyurethane hydrogel matrix on poly(vinylidene-chloride) (PVdC) foils. Dye leakage was prevented by binding the dyes to or incorporating them into microparticles. By embedding the particles in hydrogel, leakage of particles from the sensor was also prevented. The insets show transmission electron-microscopic (TEM) pictures of the sensor particles (bars are: left and right image 0.5 µm, middle image 2 µm). (B,C) Excitation (ex) (460 nm light emitting diode = LED) and emission (em) spectra of the sensor particles. Luminescence signals were optically filtered for pO2 (800/60 filter) and pH (OG530 + S8023 filters) visualization. (D,E) Detection schemes for pO2 and pH. Luminescene lifetime imaging (LLI) (D) was used for pO2-imaging. Luminescence intensities during emission were integrated in two separate time-frames, and a ratio (R) of these data gives a referenced signal for pO2. Time-domain dual lifetime referencing (tdDLR) (E) was used for pH-imaging. The combined signal intensities of pH-dependent FITC and pH-independent Ru(dpp)3 were integrated in a time-gate during excitation, and only the long-lasting reference signal of Ru(dpp)3 was detected in a time-gate during emission. The ratio (R) of these data gave a referenced signal for pH. (F,G) Calibration curves for pO2 and pH. Solving the linear fit for pO2 and the sigmoidal fit for pH enabled pO2/pHe-calculations for each pixel. Measured luminescence intensity ratios (R) obtained for each pixel were inserted in the respective equations to give referenced pO2/pHe-values, which displayed as pseudocolor maps to show the spatial distribution of these parameters.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 724

The (commercially available) polyurethane hy- remain virtually unchanged over 30 min at a resolu- drogel chosen here is a trade-off between diffusion tion of less than 5 µm [30]. Horizontal proton diffu- rates, biocompatibility, ease of handling and optical sion is minimal as seen in our spatiotemporal micro- properties. Conceivably, similar polymers may also be scopic imaging experiments [30]. Similar experiments used: other polyurethanes, hydrogels of the poly- were done for pO2, showing that oxygenation can also acrylamide, silamine, or glucosamine type, and re- be visualized at high spatiotemporal resolution [28, spective composites [33]. 31]. Another major concern is that horizontal analyte Dual visualization of pO2 and pHe on human diffusion in the hydrogel might alter the spatial reso- chronic wounds lution over time. In previous works, we studied the spatiotemporal resolution of our sensor concept [28, In humans, the vast majority of chronic wounds 30]. This is important as two diffusion processes con- are ulcers caused by chronic venous insufficiency, i.e. sistently and simultaneously occur: horizontal and reduced venous backflow from the lower limbs. vertical diffusion, both contributing to the final image. Long-standing ulcers with no tendency to reepitheli- We studied the spatiotemporal resolution using con- alize were specifically chosen (Fig. 2A, Additional tinuous microscopic luminescence imaging of a drop file 1: Table S1). Sensors were then used to visualize (pH 8) on a sensor foil impregnated in a solution of pO2 and pHe in this wound entity. pH 4. We were able to demonstrate that pH-values

Figure 2: Spatial patterns of pO2 and pHe on chronic wound surfaces. (A) Photographs of chronic venous ulcers on the lower legs of patients. Scale bars, 1 cm. (B,C) Distribution of pHe and pO2 in chronic wounds. pO2-values range from 0 up to 60 mmHg with high spatial variability. pHe-values range from ~5 up to ~8 with high interindividual variability. In large areas, pHe increased from the wound peripheries towards the wound centers. (D,I) After defining the wound margins, mean pO2/pHe for the entire wound surface was calculated (E,J,G,L). (F,K) Self-programmed computerized macros (Materials and Methods, Macro S1) served to obtain pO2/pHe-data for the different wound regions. (G,L) Mean pO2 amounted to ∼37 mmHg for the entire wound surface, and mean pHe was ∼6.8. (H) A relatively homogenous distribution of pO2-values was found when analyzing data from all wound regions. (M) There were significant differences in pHe-values between the wound periphery and regions in the wound center. A pHe-gradient was found, starting with high mean pHe in the wound center (∼7.4) and decreasing to low pHe near the wound margins (∼6.5). n = 10, mean ± s.e.m., one-way ANOVA p<0.001, *p<0.05 in post-hoc Tukey-tests.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 725

It shall be noted here that the formation of air pHe-gradients and epidermal barrier repair - a bubbles between sensor and tissue (when depositing therapeutic approach the sensor film on the wound) is detrimental because We then studied the functional impact of the it causes undesired effects: a delayed response to ox- observed pHe-variations on epidermal barrier repair. ygen, a lack of response to pH (which is wa- Using pH-adjusted culture media (Additional File 1: ter-bound), and the formation of a reservoir for oxy- Fig. S2), we first investigated pHe-effects on cell pro- gen which heavily biases accuracy. Therefore, sensors liferation. With 3H-methyl-thymidine uptake as a were gently applied to wounds starting from one marker for proliferation, we found a marked reduc- wound edge and were allowed to slowly adhere to the tion of keratinocyte proliferation at the low tissue surface by adhesion forces. The wound fluid, pHe-values found at the wound periphery when even if minimal, consistently allowed the foils to ad- compared to the high pHe at the wound center (Fig. here to the wound surface. Foils were then photoex- 3A). Moreover, we found that cell viability, as meas- cited with LEDs, and respective time-gated lumines- ured via ATP-bioluminescence assays, was critically cence signals for pO2/pHe were optically filtered and reduced at the pHe-values found at the outer wound recorded with a CCD-camera using 2D-LLI/2D- regions (Fig. 3B) as compared to the cell viability at tdDLR. We processed images with ImageJ to yield 2D the pHe found at the wound center. WST-1 tests con- pseudocolor maps of pO2/pHe in chronic wounds firmed a rise in cell viability with increasing pHe (Fig. 2B,C). (Additional File 1: Fig. S3). Therefore, keratinocyte To analyze 2D distribution of pO2 and pHe, we proliferation and viability are reduced at the pHe developed computerized macros (Additional file 5: found at chronic wound peripheries. Macro S1), capable of dividing imaging data from Next, we studied how cell migration is affected asymmetrical wound surfaces in 5 annular regions, by the pHe-values found in chronic wounds using 2D each resulting from the percental distance (necessary wound healing scratch assays (Fig. 3C) [37]. We for different wound sizes) from the wound margin. demonstrated that gap closure of keratinocyte epithe- After marking the wound margins in the photographs lial monolayers is impeded by low pHe-values (6-6.5), (Fig. 2D,I), these contours were transfered to lumi- whilst with rising pHe-values (7-7.5) the injured nescence images, and these images were subsequently monolayer closed completely after 48 h. divided into annular regions (Fig. 2E,F,J,K). Mean To further investigate the effect of pHe on mul- pO2/pHe was calculated for each region per wound tilayered epithelia, we designed 3D skin constructs from all values found in these regions. Additionally, using keratinocyte/fibroblast co-culture. After mean pO2/pHe was calculated for each entire wound wounding, a thin layer of pH-adjusted media (pHe 6.5 from all values within the margins. For details on or 7.4), which served as a surrogate for the wound image processing, see Materials and Methods. fluid, was pipetted onto injury sites in 3D skin con- Chronic wound surfaces were hypoxic at ∼37 structs. Two days after wounding, histological evalu- mmHg, with a mean pHe of ∼6.8 (Fig. 2G,L), even ation revealed a multilayered epithelium on injury though the epidermal barrier was absent in most are- sites with pHe 7.4 media, which covered the entire as. We have previously shown that one day after wound area (Fig. 3D). No wound closure was ob- wounding pHe is above 8 and pO2 is ~60 mmHg, and served in the wound models with pHe 6.5 media (Fig. that both parameters decrease during epidermal bar- 3D), demonstrating that low pHe inhibits multilayer rier restoration in physiological healing [30, 31]. Sur- epidermal repair, thus wound closure. prisingly, we found centripetally rising pHe-values in To elucidate how pHe-gradients influence chronic wounds (Fig. 2M). Despite a high variability keratinocyte migration at the single-cell level, we de- between the wounds in terms of size, shape and vised a microfluidic cell migration assay [38]. The overall level of pHe, vast areas with centripetally ris- pHe-gradients were established within the pHe-range ing pHe-values were found in the wounds. Mean measured in vivo (Fig. 3E, Additional File 1: Fig. S4). pHe-values were significantly higher with increasing In this microfluidic assay, we observed that distance from the margins (periphery-pHe ∼6.5, cen- keratinocyte motility was lower at a pHe of ~6.4 as ter-pHe ∼7.4). No corresponding pO2-gradient (Fig. compared to higher pHe-values of ~6.8 and ~7.2 (Fig. 2H) was observed, highlighting the fact that the 3 F-H, Additional File 1: Fig. S5, Movies S1-3). The pHe-gradient is not merely caused by the lack of a migration velocity of keratinocytes increased signifi- functional epidermal barrier. The degree of wound cantly at higher pHe-values (Fig. 3I,J). Furthermore, hypoxia measured here has been shown to critically the y forward migration index (yFMI) indicated that impair keratinocyte proliferation and migration in keratinocytes at low pHe of ~6.4 exhibit a low ten- healing [20, 22, 34-36]. dency to migrate towards higher pHe (Fig. 3K,L). yFMI of keratinocytes was highest at pHe ~6.8 (addi-

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 726 tional information on accumulated distance and Furthermore, we found that keratinocytes cul- endpoint y in Additional File 1: Fig. S6). As tured at the predominantly high pHe at wound cen- keratinocyte proliferation, viability, migration veloc- ters secreted markedly more IL-6 and IL-8 (important ity, and the respective yFMI are critically reduced at direct and indirect mediators of keratinocyte prolifer- the low pHe found at the wound periphery, keratino- ation and migration) [39, 40] when compared with cytes may not be sufficiently recruited from the those at lower pHe as found at the wound periphery wound margins to restore an intact epidermal barrier. (Fig. 3M, Additional File 1: Fig. S7).

Figure 3: Impact of pHe on epidermal barrier repair. All experiments: human epidermal keratinocytes (HK), (A,B,I,K,M) mean ± s.e.m. (A,B) Proliferation and viability (also see Additional File 1: Fig. S3) of HK was significantly reduced at low pHe-values. n = 12, NS = not significant, one-way ANOVA p<0.001 and *p<0.05 in post-hoc Holm-Sidak test. (C) In a 2D scratch assay, there was almost no movement of the HK monolayer at low pHe of 6.0, whereas with rising pHe centripetal monolayer movement increased. Black dashed lines = migration front after injury, red dashed lines = migration front after the respective time, white arrows = regions of significant migration. E = epithelium, I = injury, n = 3. (D) 3D Skin constructs of co-cultured HK and fibroblasts. After mechanical injury, healing was studied with pH-adjusted media (as substitutes for the wound fluid) at the injury site. No healing was observed at low pHe 6.5 after 48 h, whereas at pHe 7.4 the wound was covered with a multilayered epithelium. Scale bars, 200 µm, D/E = dermis/epidermis equivalent, n = 3. (E) Chamber of a microfluidic cell migration assay. Pseudocolor scale and actual pHe-gradient (horizontally downscaled to show the gradient over the entire chamber) in the microfluidic chamber (visualization via luminescent BCECF). Numbers 1-5 denote regions for further analyses. HK (at 0 h) and according migration traces (colored) within 16 h are shown. Scale bar, 200 µm. (F-H) Trajectory lines of HK superimposed on coordinates (start at x/y = 0 µm). Mean slow cell velocity ≤ 12.6 µm/h, mean fast cell velocity > 12.6 µm/h. At low pHe, nearly all cells exhibited slow movement, whereas migration distance and the proportion of fast cells increased with rising pHe (E-H, results from one representative experiment, additional experiments/videos: Additional File 1: Fig. S5, Movies S1-3). n = 3 experiments. (I-L) Results from three microfluidic assays. Migration velocity (I) increased with rising pHe. The corresponding scatter point plot (J) shows results for all cells (red lines = medians). (K,L) y forward migration index (yFMI) was low at pHe ~6.4, and highest at pHe ~6.8. (L) red lines = medians. Together with velocity results, this demonstrated minimal chemotactic/migration activity at low pHe as found at wound peripheries. Cells tracked in 3 experiments: region 1 (n = 42), 2 (n = 60), 3 (n = 65), 4 (n = 69), 5 (n = 44). a.u. = arbitrary units, one-way ANOVA on ranks p<0.001 and *p<0.05 in post-hoc Dunn´s test. (M) Secretion of mediators of proliferation/migration, namely IL-6 and IL-8, is markedly increased at high pHe (see Additional File 1: Fig. S7). n = 3, one-way ANOVA p<0.001 and *p<0.05 in post-hoc Tukey test.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 727

Proton transporter expression in chronic 4A,B, Additional File 1: Fig. S8) revealed an increase wounds in total and plasma membrane NHE1-density at wound peripheries, yet not at the central wound re- To decipher how the pHe-gradients in chronic gion, when compared with surrounding intact tissue. wounds might arise, we studied the expression levels In the core of the wound, NHE1 protein-expression of essential pHe-regulatory ion transporters in punch seemed to decrease strongly again, reaching a level biopsies from human wounds (Additional file 1: Ta- similar to that in the normal tissue. For validation, ble S2). We assessed the protein expression level of quantitative reverse transcription PCR (qRT-PCR) for Na+/H+-exchanger-1 (NHE1), a ubiquitously ex- NHE1 was done (Fig. 4C), demonstrating a similar pressed transporter which is highly regulated (e.g. by pattern of mRNA-expression, with greatly increased cytokines, H2O2, reduced intracellular pH) [41, 42], NHE1-mRNA-levels at the wound periphery, de- and which contributes importantly to the develop- creasing again at the wound center. Western blots ment of acidic pHe, e.g. in carcinomas [41, 43, 44]. confirmed the presence of a single band of the correct Immunofluorescence analysis of NHE1 in sections size for each transporter. from intact skin, wound peripheries and centers (Fig.

Figure 4: Expression and localization of pHe-regulatory proton transporters in chronic wounds. (A-F) Three biopsies (adjacent intact skin, wound periphery, wound center) were taken from each patient (n = 3) and analyzed for proton transporter expression. The immunofluorescence images shown are representative of at least three different sets of images from each wound region. Sections were counterstained with DAPI (blue) for nuclei staining. Scale bars, 20 µm. (A-C) Representative (A,B) immuno- fluorescence staining and (C) quantitative reverse transcription PCR (qRT-PCR) of Na+/H+ exchanger-1 (NHE1) from adjacent intact skin, peripheral and central wound tissues from chronic wound patients. (A,B) NHE1-expression is strongly upregulated in peripheral wound tissue, but not in the central part of the wound. NHE1-antibodies were tested in Western blotting and found to produce bands of the correct size. Additional immunofluorescence images are shown in Additional File 1: Fig. S8. (C) mRNA of NHE1 was extracted from adjacent normal tissue, the wound peripheries and wound centers. We found that NHE1-mRNA was strongly upregulated at the wound periphery as compared to intact skin and the wound center. Mean ± s.e.m., n = 3. (D-F) The expression levels of the (D) plasma membrane vacuolar-type ATPases subunit B2 (ATP6V1B2) and the (E) (H+-lactate) monocarboxylate cotransporter 1 (MCT1) were not significantly upregulated in the peripheral and central wound sections as compared to intact skin. Additional immunofluorescence images are shown in Additional File 1: Fig. S9. Taken together, these results show that NHE1-levels are centrifugally upregulated in chronic wounds, and suggest that increased NHE1-mediated hydrogen ion extrusion acidifies the wound fluid with increasing distance from the center.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 728

To assess whether other pHe-regulatory trans- late VEGF production of keratinocytes [53], and VEGF porters might exhibit similar gradients of expression, is a key molecule in neoangiogenesis during healing. we performed immunofluorescence analysis for the However, at the low pHe as found at the wound pe- (H+-lactate) monocarboxylate transporter 1 (MCT1), riphery, we found low levels of INFγ in keratinocyte and for the vacuolar ATPase subunit B2 (ATP6V1B2). supernatants. Therefore, the pHe-gradients found are The protein levels of both transporters were virtually possibly detrimental for neoangiogenesis at the identical in intact skin, wound peripheries and wound wound periphery. More experiments have to be done centers (Fig. 4D-F, Additional File 1: Fig. S9). Taken to study the effects of the pHe-generated cytokine and together, these data are highly consistent with the chemokine secretion patterns on cell migration and observed pHe-gradient: NHE1-expression is upregu- proliferation. Additionally, the activity of MMPs in- lated at the wound periphery, and increased volved in healing [5] is regulated by pHe, and it has NHE1-mediated hydrogen ion extrusion acidifies the yet to be shown how the varying pHe-values reported wound fluid with increasing distance from the center. here affect these enzymes in the different wound re- It is, however, impossible to add an NHE1-inhibitor gions. (e.g. 5´N-ethylisopropylamiloride = EIPA, cariporide) Next, we showed that the pHe-gradients found prior to measurements on humans to block on chronic wound surfaces are predominantly gener- NHE1-mediated acidification of the wound fluid. ated via centrifugally increasing NHE1-expression. NHE1-expression is strongly upregulated at the Discussion wound periphery, and increased NHE1-mediated Gradient sensing is an essential biological hydrogen ion extrusion acidifies the wound fluid in mechanism governing directional cell migration in this region. Other acid extruding transporters, such as both prokaryote and eukaryote cells [45-50]. In phys- MCT1 and vacuolar type H+-ATPases, were not sig- iological wound healing, gradients of growth factors nificantly upregulated at the wound periphery. The (e.g. EGF) [8, 11], cytokines and chemokines [8], or immunofluorescence and qRT-PCR data in conjunc- small signaling molecules (e.g. H2O2) [10] have been tion with the measured pHe-gradients suggest that shown to be decisive cues in mediating barrier repair. NHE1 is predominant, in accordance with its major In this study, we showed that gradients of ex- physiological roles in acid extrusion. It is obviously tracellular hydrogen ion concentrations exist on the impossible to add an NHE1-inhibitor (EIPA or cari- surface of chronic wounds, and that these poride) prior to measurements on humans. As NHE1 pHe-gradients negatively affect keratinocyte viability, is the only NHE-isoform found in keratinocytes and proliferation and migration at the wound periphery. epidermis [54], no other isoforms are relevant here. Decreasing acidity towards the wound center inhibits With respect to MCT-isoforms, no studies exist spe- centripetal keratinocyte recruitment, thus preventing cifically addressing their roles in keratino- wound closure as shown in 2D and 3D wound healing cytes/epidermis. MCT1 is, however, the most wide- assays as well as in microfluidic cell migration ex- spread isoform, with MCT2, 3, and 4 showing much periments (scheme in Additional File 1: Fig. S10). more restricted expression patterns. MCT2-expression It has been reported recently for αvβ3 CHO-B2 is very low in human tissues (with the exception of cells and microvascular endothelial cells, that low pHe testis), MCT3 is restricted to retina and the choroid leads to increased integrin αvβ3-expression and sub- plexus, and MCT4-expression is generally limited to sequently reduced cell migration velocity [51, 52]. tissues which use glycolysis for their energy metabo- Based on our findings, this means that, at the wound lism (e.g. white skeletal muscle fibers, astrocytes, periphery, these cells would be less motile and more white blood cells, chondrocytes, certain tumor cells, prone to maintain their adherence, exactly at the re- cells under severe hypoxia) [55]. Therefore, it can be gion from which they are recruited from for healing. concluded that no other NHEs than NHE1 contribute We also detected high levels of direct and indi- to the pHe-gradient, and that a major contribution rect mediators of keratinocyte proliferation and mi- from MCT2-4 seems very unlikely. gration, namely IL-6 and IL-8 [39, 40], at the pHe Further, we describe, for the first time, the visu- found at the wound center (Additional File 1: Fig. alization of chronic wound oxygenation, and we S7). In contrast, keratinocytes secreted low amounts found marked hypoxia (∼37 mmHg), albeit with no of IL-6 and IL-8 at the low pHe-values found at the correlating pO2-gradient. During physiological heal- wound periphery, which may also contribute to re- ing, we have previously shown that pO2 amounts to duced centripetal keratinocyte recruitment. Another ∼60 mmHg on the first day after epidermal barrier interesting finding was that interferon-γ (INFγ) re- removal, and that pO2 gradually decreases during lease from keratinocytes increased with rising pHe epidermal barrier recovery [31]. Interestingly, pO2 in (Additional File 1: Fig. S7). INFγ is known to stimu- chronic wounds is markedly lower, despite the lack of

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 729 epidermal coverage. Such low pO2-values critically pHe-dependent cell proliferation and migration in impair keratinocyte proliferation [19, 20, 36] (partially wounds and tumors [43, 44, 59, 62]. by miRNA 210 upregulation) [34] and protein syn- thesis (e.g. cytokines, chemokines, growth factors), Materials and methods both processes being prerequisites for healing. Re- Luminescent Microparticles cently, it was also shown that hypoxia may partially arrest the cell cycle in keratinocytes via hypoxia in- The oxygen probe PdTPTBP (λex 444 nm, λem 797 ducible factor-1α (HIF-1α), which leads to miRNA nm, lifetime 50-300 µs depending on pO2, Sig- 210-induced silencing of the transcription factor E2F3. ma-Aldrich, Seelze, Germany) was incorporated into HIF-1α is, however, also crucial for keratinocyte mi- oxygen-permeable PSAN (30 wt% acrylonitrile con- gration via regulation of laminin-332 [35]. Other ef- tent, Mw 185,000, Sigma-Aldrich) in a 1/50 ratio to fects of hypoxia are mediated by miRNA 21, which form oxygen sensor beads (PdTPTBP-PSAN, diameter silences early growth response factor 3 (EGR3). Ac- 0.5-1 µm). Polymer and dye were dissolved in dime- thylformamide (0.12 mL per mg net weight, 60 °C). cording to the literature [19], the pO2 of ∼37 mmHg measured in this study may reduce keratinocyte pro- After cooling the mix to room temperature, particles liferation by ∼60% when compared to normoxic con- precipitated under stirring (350 rpm) with the addi- ditions. Hypoxia has, however, also been reported to tion of distilled water (0.32 mL per mg net weight, -1 be a stimulus for keratinocyte migration, yet most drop-wise, 1 mL s ). The solution was stirred for 30 patients with chronic wounds are of advanced age, minutes to ensure complete precipitation. The reac- and it is known that hypoxia loses its potential as a tion can be scaled up to generate 250 mg of polymer. stimulus for cell migration with increasing patient age Particles were washed (6 x with ethanol, 8 x with dis- [21, 56]. Therefore, we conclude that the low tilled water), and filtered via centrifugation (10 min, 1,030 x g) after each washing step. Reactions were pO2-values we measured on chronic wound surfaces contribute to reduced keratinocyte proliferation and conducted at room temperature unless otherwise migration, thus impairing barrier rapair. Other stated and particles were freeze-dried for storage miRNAs, such as miRNA-198, have recently been (Modulyo, IMA Edwards, Bologna, ). shown to be decisive factors in chronic wound healing pH-dependent FITC (λex 495 nm, λem 525 nm, [57]. lifetime < 5 ns, Sigma-Aldrich) was conjugated to AC In summary, we present the first observation of particles (PreSens, Regensburg, Germany) in a 1/50 wt/wt ratio in sodium bicarbonate buffer (50 mM, pH presumably NHE1-generated pHe-gradients on chronic wounds. We further provide unique evidence 9, 1.8 mL per mg FITC) to form FITC-AC particles (2 h reaction time, diameter 0.75-3 µm). Residual amino for the notion that pHe-gradients inhibit keratinocyte supply from the wound margins in chronic wounds groups on the particle surface were blocked with ace- tic anhydride (Sigma-Aldrich). FITC-AC was reacted (Additional File 1: Fig. S10), a previously un- with acetic anhydride and ethanol in a 1/2 wt/wt described pathomechanism. Modulating pHe in chronic wounds poses a promising and feasible ratio (0.1 mL per mg AC, 12 h reaction time). Particles strategy for future therapeutics (e.g. dynamic hydro- were washed (8 x with distilled water), and filtered gels) [58] to improve patient outcomes. Hydrogels via centrifugation (10 min, 1,030 x g, EDA12, Hettich, Tuttlingen, Germany) after each washing step. which restore a higher pHe in chronic wounds, espe- ∼ cially at the periphery, might stimulate increased Ru(dpp)3 (λex 441 nm, λem 597 nm, lifetime 6 µs, proliferation and centripetal migration of keratino- Sigma-Aldrich) was incorporated in PAN (Sig- cytes. pH-gradients also play an important role in ma-Aldrich) at a 1/50 ratio to form Ru(dpp)3-PAN tumor metabolism [43, 44], and therefore these find- particles (diameter 0.5-1 µm). Particles were formed ings provide evidence for a novel link between cancer by precipitating dye and polymer dissolved in dime- thylformamide (0.2 mL per mg net weight) with dis- and wounds in terms of pHe-regulated cell prolifera- tion and migration [59]. In humans, there are four tilled water (0.28 mL per mg net weight, drop-wise -1 proton-sensing G protein-coupled receptors (GPCRs), addition, 1 mL s ) and the subsequent addition of which are possibly involved in sensing the Brine (0.08 mL per mg net weight). The reaction can be scaled up to 250 mg of polymer. Particles were above-mentioned pHe-gradients: G2A (GPR132), GPR4, OGR1 (GPR68), and TDAG8 (GPR65) [60, 61]. washed (8 x with distilled water, 4 x with ethanol), and filtered after each washing step via centrifugation These GPCRs are mainly activated when pHe drops (10 min, 1,030 x g). below the physiological range, i.e. at the pHe-values we found at the wound periphery. Further studies Particle sizes were assessed with a LEO912 AB should elucidate the role of proton sensing GPCRs in transmission electron microscope (Carl Zeiss, Ober- kochen, Germany). The microparticle design effi-

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 730 ciently prevents Förster resonance energy transfer current revision of the Declaration of Helsinki. Im- (FRET). paired blood circulation (venous backflow, arterial inflow, microcirculation) is often a precursor to the Sensor Foils development of chronic wounds. Study subjects were The microparticles (PdTPTBP-PSAN, FITC-AC, included based on the main causative disease being Ru(dpp)3-PAN) were mixed (ratio 2/3/1). The parti- venous insufficiency, one wound diameter being at cle mix was dispersed (15 mg mL-1) in hydrogel solu- least 3 cm (for adequte analysis of signal distribution tion consisting of 5% wt/vol polyurethane hydrogel in an in vivo setting), and the absence of any healing (type D4, Cardiotech International Inc., Plymouth, tendency. MN, USA) in ethanol/water (90/10 vol/vol). This mixture was spread on a transparent PVdC-foil In Vivo Imaging (thickness 25 µm, Saran plastic wrap, S.C. Johnson & Sensor foils were preconditioned (30 min, room Son Inc., Racine, WI) with a knife-coating device (K temperature) in Ringer´s solution (B. Braun Mel- Control Coater model 101, RK Print-Coat Instruments sungen, Melsungen, Germany) before use. Sensors Ltd., Litlington, UK) to form a 120-μm-thick film (Fig. were gently applied to the wound surface starting 1A). After drying, sensor layer thickness was 6 μm. from one margin of the wound. Adhesion forces ena- One mL of sensor mix generates a sensor film of 83 bled the foils to adapt to tissue surfaces. The for- cm2. mation of air bubbles between sensor and tissue was thereby prevented and uniform contact ensured. We Luminescence Imaging applied no pressure to the foils during measurements. Imaging for pO2/pHe was undertaken using a The sensors remained on the wounds for 5-10 minutes time-gated imaging setup (ImageX TGI System, Pho- and covered the entire wound surface and parts of the tonic Research Systems, Newhaven, UK) with an in- surrounding tissue. The distance between sensor and tegrated 12-bit CCD chip (640 x 480 pixels). The cam- camera was focus-controlled (25 cm). All images were era was combined with an array of light emitting di- recorded >1 min after sensor application to yield suf- odes (LED, λ = 460 nm, Luxeon V Star LXHL-LB5C ficient sensor response (oxygen and proton diffusion). 5W, Philips Lumileds Lighting Company, San Jose, After recording oxygen images, the filters of the cam- CA, USA) equipped with a FITC excitation filter era were changed and pHe-measurements were per- (Chroma Optical, Brattleboro, VT, USA) (Fig. 1B). formed. The LLI-scheme (Fig. 1D,F) was used for Image Processing pO2-imaging (parameters: subtract background, 70 μs gate width, 1 μs delay time for window 1 and 36 µs For pO2/pHe raw data processing, integrated delay time for window 2, 500 μs lamp pulse, 400 Hz luminescence intensities in gate 1 were divided by the recording frequency, 2 s integration time). An 800/60 according intensities in gate 2 (ImageX software, pro- nm bandpass filter (Omega Optical, Brattleboro, VT, vided with the TGI camera system, Photonic Research USA) served to separate emitted oxygen probe lumi- Systems). pO2/pHe raw data were stored as 16-bit nescence from background and pHe-dependent sig- black-and-white pictures (TIFF-format). nals (Fig. 1C). Triplicates of each image were collected The following processing steps were performed in every measurement cycle. for each wound with Photoshop CS4 (Adobe Systems, The tdDLR-scheme (Fig. 1E,G) was used for San Jose, CA, USA): pO2/pHe raw data and a photo- imaging pHe-distribution (parameters: subtract back- graph of the wound were imported into one 16-bit ground, 5 μs gate-width, 5.5 µs delay time for gate 1, PSD-file. The original wound picture was converted 0.25 μs delay time for gate 2, 6 μs lamp pulse, 400 Hz to grayscale and then duplicated. Layers were re- recording frequency, 100 ms integration time). The named: “Pseudocolor pH”, “Pseudocolor pO2”, glass filters OG530 and S8023 (Schott, Mainz, Ger- “Photo”, and “Original”. “Pseudocolor pH” and many) were used as emission filters (Fig. 1C). Tripli- “Pseudocolor pO2” were superimposed onto the cates of each image were recorded in each measure- identical real-color wound pictures (“Photo” and ment cycle. “Original”) using the transformation tool. The wound margins were defined in the “Photo” layers, and re- Study Subjects gions outside the respective margins were erased. The Participants were provided with verbal and layer “Photo” was selected and the action “macro for written information on the study, and signed in- preparation” (Macro S1) was started. Thereby, the formed consent was obtained from each participant. size of the image was adapted and standardized, and The local ethics committee gave approval (University the defined wound area was transferred to the raw Medical Center Regensburg, No. 06/171: 2007), and data pictures. Subsequently, the pellucid region of the experiments were conducted in accordance with the

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 731 layer “Photo” was selected (“magic wand tool”) and pH-adjusted media the action “macro for slicing” (Macro S1) was exe- In order to maintain experimentally induced cuted. During this action, raw data were sliced into variations in pHe, we modified R&G media such that five consecutive regions, starting from the wound DMEM was substituted with Hank´s buffered M199 periphery towards the center. Next, the layer “Pseu- media (Invitrogen) with reduced bicarbonate. Whilst docolor pH” was selected and the action “macro for maintaining a similar salt concentration, this substi- measurement” ( ) was executed. Macro S1 tution allowed a range of pHe to be stably maintained The grayscale mean value of the in ambient CO2 (also see Additional File 1: Fig. S2). pHe/pO2-images from the total wound surface as well Media was adjusted to pH 6, 6.5, 7, 7.5 and 8 with the as from each slice were recorded using the “meas- dropwise addition of 1M HCl or 2M NaOH. urement tool”. The “measurement log” that buffers the data was exported to a CSV-file and imported into 3H-methyl-thymidine uptake, ATP-assays and Excel (Microsoft, Redmond, WA, USA). The imported WST-1 tests CSV-data were multiplied by a factor of 0.0022 for Keratinocyte proliferation was assessed using a re-conversion to raw data values. These values were tritiated thymidine uptake assay. Cultured keratino- converted to respective pO2/pHe-values via the cali- cytes were seeded and exposed to pH-adjusted media bration equations (Fig. 1F,G). as above. Keratinocytes were incubated with 0.5 µCi Pseudocolor image processing was done with well-1 3H-Methyl-thymidine (PerkinElmer, Beacons- ImageJ (http://rsbweb.nih.gov/ij/). For illustration pur- field, UK) for a further 24 hours, lysed using 0.1% SDS poses, pseudocolor maps for pO2/pHe on wound and 0.1 M NaOH, harvested onto a filter mat using a surfaces were superimposed to photographs using Filtermate cell harvester (PerkinElmer, UK), and ra- Adobe Photoshop (Fig. 2A-C). dioactivity was measured using a MicroBeta counter Keratinocyte culture for proliferation and (PerkinElmer, UK) following the addition of scintil- viability lant. Keratinocyte viability after 24 hours at pHe 6, 3 For H-methyl-thymidine uptake and 6.5, 7, 7.5 and 8 (n = 12 each) was measured using an ATP-bioluminescence assays, primary human ATP-assay. Keratinocytes were seeded into 96-well keratinocytes were isolated and cultured as described plates (5,000 cells well-1) and allowed to attach for 24 previously [63]. Discarded skin was obtained during hours in standard R&G growth media before remov- routine surgical procedures (UK NRES approval: ing the media and washing off any unattached cells. 06/Q1907/81) with the patients´ informed consent. Cells were incubated for 24 hours in pH-adjusted The epidermis and dermis were separated by diges- media at ambient CO2 prior to measurement of -1 tion with 4 mg ml dispase (10 min, Invitrogen, Life ATP-activity using an ATPlite-assay (PerkinElmer) Technologies Ltd., Paisley, UK) prior to liberation of and measurement of luminescence using a Mi- keratinocytes from the epidermis by incubation with croBeta-counter (PerkinElmer). 0.5% trypsin (30 min, Invitrogen). Isolated keratino- Cell vitality was additionally assessed with the cytes were co-cultured with a feeder-layer of lethally WST-1 cell proliferation reagent (Roche, , irradiated 3T3-cells in standard Rheinwald and Green Germany). Viable cells reduce the WST-1 reagent to a media (R&G) comprising DMEM and Ham’s F12 me- water-soluble formazan dye. The absorbance at 450 dium (Invitrogen) at a 3:1 ratio, supplemented with nm correlates to the vitality of the cells. Briefly, adult 10% FCS (Gibco, Life Technologies Ltd., Paisley, UK), human keratinocytes (5,000 cells cm-2, passage 3) were -1 10 ng mL human recombinant EGF (Invitrogen), 10 seeded in 96-well plates. After 24 hours, the medium nM cholera toxin (Sigma Aldrich, Gillingham, UK) was changed to the pH-adjusted media and cells were -1 and 0.4 mg mL hydrocortisone (Sigma-Aldrich, UK). incubated for 24 or 48 h, respectively. Subsequently, For water-soluble tetrazolium-tests (WST-1), WST-reagent was added to the wells for two hours adult human epidermal keratinocytes (HEKa, Gibco) and the absorbance at 450 nm was measured using a were cultured in EpiLife medium supplemented with TECAN infinite F200 plate reader (TECAN, Crail- human keratinocyte growth supplement (HKGS, sheim, Germany). Experiments were undertaken with Gibco) and a Pen/Strep solution in a humidified at- six replicates per treatment. mosphere (37°C, 5% CO2). After reaching sub-confluency, cells were passaged using a 0.25% 2D and 3D wound healing models trypsin solution. For viability assessment and 2D mi- For 2D wound healing assays, wells of a 24-well gration assays, cells in passage 3 were used. plate were incubated with 500 µl of a 10 µg ml-1 fi- bronectin/PBS (Sigma-Aldrich) solution (10 µg cm-2) overnight. Afterwards, adult human keratinocytes

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 732

(passage 3) were seeded into wells (5,000 cells cm-2). tive (UPLFLN4XPH/0.13, Olympus, Hamburg, Ger- At confluency, we scratched the cell monolayer using many). Phase-contrast images were taken every 2 min. a pipette tip and the medium was changed to the Fluorescence images (λex 434/17 nm and 492/18 nm, pH-adjusted media. Pictures of the monolayer λem 530/20 nm) were taken alternately at 1 or 30 movement were taken after 0, 12, 24, and 48 h. Ex- frames h-1 and at 10 or 300 ms exposure time, respec- periments were undertaken in triplicate. tively. Cell migration data analysis was performed For 3D wound healing assays, fibroblasts, FCS using ImageJ software (http://rsbweb.nih.gov/ij/) and bovine collagen I (Nutacon, Leimuiden, Nether- including the plug-ins Manual Tracking (F. Corde- lands) were brought to neutral pH and poured into liers, Institut Curie, Orsay, ) for imaging sin- 3D culture filter inserts for 6-well plates with a growth gle-cell migration tracks, and Chemotaxis Tool (ibidi, area of 4.2 cm2 (BD Biosciences, , Germa- Martinsried, Germany) to calculate migration veloci- ny). After cultivation for several hours at 37°C, ty, yFMI, accumulated distance, and endpoint y. We pH-adjusted medium was added and the system was used Daniel´s XL toolbox (http://xltoolbox. transferred to an incubator (5% CO2, 95% humidity). sourceforge.net) plug-in for Microsoft Excel (Mi- After a second incubation, primary human keratino- crosoft Corporation, Redmond, WA, USA) to spread cytes were added on top of the collagen matrix. On scatter plots. The Cell^R image processing software the next day, the construct was lifted to the air-liquid (OlympusSIS, Muenster, Germany) was used for ra- interface and medium was changed to a differentia- tiometric analysis of BCECF fluorescence images. tion medium [64]. After 2 days, we induced a wound ELISAs by scratching the epidermal layer with a needle. The injured constructs were incubated with pH-defined Adult human epidermal keratinocytes (Gibco) media (pHe 6.5 and 7.4). To mimic in vivo wound were cultured in triplicates under different pHe. 150 µl healing more closely, 50 µl of the media (simulating supernatants of cell culture were collected after 48 h. wound fluid) were applied onto the epidermal layer. Samples were tested using a 16-Plex human cytokine After 48 h, the skin constructs were punched, em- array (Quansys, Logan, UT, USA) according to man- bedded in tissue freezing medium and kept at -80°C ufacturer´s protocols. Chemiluminescence images overnight. Subsequently, the skin constructs were cut were acquired by BioRad Versa Doc 4000 MP work- vertically into slices (10 µm thickness, Frigocut 2800 station (Bio-Rad, Munich, Germany) using a gain of N, Leica, Bensheim, Germany) and stained with con- 2X, binning of 1X1 and auto exposure time settings. ventional haematoxylin and eosin (H&E). Images were exported as 16 bit TIFF raw data for quantitative analysis using the Q-View software Microfluidic cell migration assays (Quansys). Human primary epidermal foreskin keratino- Tissue harvesting cytes were cultivated (37 °C, 5% CO2) in EpiLife Me- dium supplied with 1:100 human keratinocyte growth Punch biopsies from wound margins, wound supplement (HKGS, S0015, Life Technologies, Darm- centers, and intact surrounding tissue were taken stadt, Germany). Cells were passaged every 2-3 days from patients with chronic venous wounds (ulcers). and used for migration experiments at passage 4 or Wound samples were cut in half along a central line to earlier. Microfluidic channels (µ-Slides Chemotaxis allow sufficient specimens for immunohistochemistry 3D, ibidi, Martinsried, Germany) were coated with 70 (IHC), quantitative reverse transcription PCR µg ml-1 Fibronectin (Sigma, Germany), dried, and (qRT-PCR) and Western blots. Allprotect Tissue Rea- filled with 2-3 x 106 cells ml-1. The cells adhered for 2-4 gent (Qiagen, Germantown, MD, USA) was used for h before the pHe-gradient was established by filling preservation. the two feeding reservoirs of the µ-Slides with Me- Histology and Immunofluorescence dium 199 (Life Technologies, , Germany) Chronic wound biopsies were fixed in para- adjusted to pHe 7.4 and 6.4, which was supplemented with 1:100 HKGS and 5 or 50 µM formaldehyde and routinely processed for paraffin 2´,7´-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescei embedding. After blocking with 5% BSA in phosphate n (BCECF, Life Technologies, Darmstadt, Germany) buffered saline supplemented with 0.1% Tween 20, sections were incubated with primary antibodies for pHe-visualization. (NHE1 and ATP6V1B1/B2: Santa Cruz Biotechnolo- Cell migration and pHe-gradient stability was gy; MCT1: Millipore, Billerica, MA, USA) at room monitored at 37 °C and ambient CO2 conditions for a time period of 14 or 16 h by using a fully automated temperature for 1 h followed by several washes in PBS microscope system (Cell^R, Olympus, Hamburg, and incubation with fluorescently-labelled secondary Germany) equipped with a 4 x phase-contrast objec- antibodies, namely Alexa Fluor 594 goat anti- rabbit IgG (H+L) and Alexa Fluor 488 goat anti-mouse

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 733

IgG (H+L) (Invitrogen). Images were acquired using pHe-values in the 5 defined wound regions. One-way an Olympus Bx63 epifluorescence microscope ANOVA and post-hoc Holm-Sidak testing were (100x/1.4 NA objective), and image processing (in- conducted to assess for statistical differences in pro- tensity adjustment only, quantitatively equal for all liferation and viability of keratinocytes at varying samples) was performed in Adobe Photoshop (Adobe pHe. To check for differences in migration velocity, Systems). y-FMI, directionality, and endpoint y, we performed one-way ANOVA on ranks and post-hoc Dunn´s RNA and Protein Isolation testing. Multiplex ELISA results were analyzed using Total RNA and proteins were isolated from one-way ANOVA and post-hoc Tukey-tests or wound biopsies from human patients using the RNe- one-way ANOVA on ranks and Dunn´s post-hoc test. asy Mini Kit (Qiagen) according to the manufacturer's To detect differences in qPCR data, one-way ANOVA instructions. RNA aliquots were stored in a -80 oC and post-hoc Tukey tests were used. freezer and extracted protein was stored at -20 oC. Supplementary Material Quantitative Reverse Transcription PCR (qRT-PCR) Additional File 1: Figures S1-S10. Tables S1-S2. Still Image for Movies. For gene expression analysis, isolated total RNA http://www.thno.org/v04p0721s1.pdf was reverse-transcribed using Sensiscript RT Additional File 2: Kit (Qiagen), and cDNA-transcripts were amplified Movie S1. by qPCR using SYBR Green (Applied Biosystems, http://www.thno.org/v04p0721s2.avi Cheshire, UK) and primers designed using Pri- Additional File 3: mer-BLAST from NCBI. Triplicate samples were used. Movie S2. Expression levels of mRNA were evaluated using the http://www.thno.org/v04p0721s3.avi comparative threshold cycle (Ct) method as normal- Additional File 4: ized to β-actin. Relative expression levels of mRNA Movie S3. were calculated between tissue sections. http://www.thno.org/v04p0721s4.avi Western Blots Additional File 5: Macro S1. To assess protein expression, proteins isolated http://www.thno.org/v04p0721s5.atn from each patient’s wound section were fractionated by 10% sodium dodecyl sulfate-polyacrylamide gel Acknowledgments electrophoresis (SDS-PAGE) and transferred to a polyvinylidene difluoride (PVDF) membrane. Blots The work was supported by grants of the Ger- were then probed with respective primary antibodies man Research Foundation (DFG, BA 3410/3-1, BA specific for each protein: β-actin (A5316 from Sig- 3410/4-1, SCHA 1009/7-1, WO 669/9-1), and the Re- ma–Aldrich; NHE1, MCT-1 and ATP6V1B1/B2 from ForM-B program (S.S., University Medical Center Santa Cruz Biotechnology, CA, USA). Dilutions were Regensburg, Regensburg, Germany). The work was made according to the manufacturer’s recommenda- supported by the German Research Foundation DFG tion. After incubation with respective primary anti- within the funding program open access publishing. bodies, the membranes were washed in tris-buffered We thank Beate Morgenstern (Fraunhofer IBMT, saline supplemented with 0.1% Tween 20 and incu- Potsdam, Germany) for cell culture assistance and bated with the appropriate secondary antibodies Heiko Siegmund (Center for Electron Microscopy, (Sigma-Aldrich) followed by development with the Institute of Pathology, University Medical Center BCIP/NBT phosphatase substrate system (Pierce Bi- Regensburg, Regensburg, Germany) for his work on otechnology, Rockford, IL). β-actin served as loading transmission electron microscopic imaging. control. Author contributions Statistics S.S., R.J.M., O.S.W., and P.B. designed research. We used Sigma Plot 11.0 (Systat Software Inc., S.S. conceived functional experiments. S.S., R.J.M., Chicago, IL, USA) for all analyses. Data are given as M.K., S.C.K., S.G., O.F., S.K., J.R.S., K.W., K.T.W., mean ± standard error of the mean (s.e.m.) except M.A., U.S., and J.S. performed experiments. S.S., otherwise denoted. We considered a p-value below R.J.M., M.K., O.F., J.R.S, S.F.P., and P.B. analyzed the 0.05 as significant. Results were marked as significant data. C.M., L.P., C.D., M.G., M.B., O.S.W., and M.L. with asterisks within the graphs. helped to analyze data. S.S., O.S.W. and P.B. obtained We utilized one-way ANOVA and post-hoc grant support. R.J.M., M.K., S.C.K., S.G., O.F., S.K., Tukey-testing to analyze differences between mean J.R.S., S.F.P., and P.B. wrote parts of the manuscript.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 734

S.S. compiled and wrote the final manuscript. 30. Schreml S, Meier RJ, Wolfbeis OS, Landthaler M, Szeimies RM, Babilas P. 2D luminescence imaging of pH in vivo. Proc Natl Acad Sci U S A. 2011; 108: 2432-7. Competing Interests 31. Schreml S, Meier RJ, Wolfbeis OS, Maisch T, Szeimies RM, Landthaler M, et al. 2D luminescence imaging of physiological wound oxygenation. Exp The authors have declared that no competing Dermatol. 2011; 20: 550-4. 32. Wang XD, Wolfbeis OS. Optical methods for sensing and imaging of oxygen: interest exists. materials, spectroscopies and applications. Chem Soc Rev. 2014; [Epub ahead of print]. 33. Mateescu A, Wang Y, Dostalek J, Jonas U. Thin hydrogel films for optical References biosensor applications. Membranes. 2012; 2: 40-69. 1. Gurtner GC, Werner S, Barrandon Y, Longaker MT. Wound repair and 34. Biswas S, Roy S, Banerjee J, Hussain SR, Khanna S, Meenakshisundaram G, et regeneration. Nature. 2008; 453: 314-21. al. Hypoxia inducible microRNA 210 attenuates keratinocyte proliferation 2. Jacinto A, Martinez-Arias A, Martin P. Mechanisms of epithelial fusion and and impairs closure in a murine model of ischemic wounds. Proc Natl Acad repair. Nat Cell Biol. 2001; 3: E117-23. Sci U S A. 2010; 107: 6976-81. 3. Singer AJ, Clark RA. Cutaneous wound healing. N Engl J Med. 1999; 341: 35. Fitsialos G, Bourget I, Augier S, Ginouves A, Rezzonico R, Odorisio T, et al. 738-46. HIF1 transcription factor regulates laminin-332 expression and keratinocyte 4. Martin P. Wound healing - aiming for perfect skin regeneration. Science. 1997; migration. J Cell Sci. 2008; 121: 2992-3001. 276: 75-81. 36. Straseski JA, Gibson AL, Thomas-Virnig CL, Allen-Hoffmann BL. Oxygen 5. Martins VL, Caley M, O'Toole EA. Matrix metalloproteinases and epidermal deprivation inhibits basal keratinocyte proliferation in a model of human skin wound repair. Cell Tissue Res. 2012. and induces regio-specific changes in the distribution of epidermal adherens 6. Bickers DR, Lim HW, Margolis D, Weinstock MA, Goodman C, Faulkner E, et junction proteins, aquaporin-3, and glycogen. Wound Repair Regen. 2009; 17: al. The burden of skin diseases: 2004 a joint project of the American Academy 606-16. of Dermatology Association and the Society for Investigative Dermatology. J 37. Liang CC, Park AY, Guan JL. In vitro scratch assay: a convenient and Am Acad Dermatol. 2006; 55: 490-500. inexpensive method for analysis of cell migration in vitro. Nat Protoc. 2007; 2: 7. Sen CK, Gordillo GM, Roy S, Kirsner R, Lambert L, Hunt TK, et al. Human 329-33. skin wounds: a major and snowballing threat to public health and the 38. Zengel P, Nguyen-Hoang A, Schildhammer C, Zantl R, Kahl V, Horn E. economy. Wound Repair Regen. 2009; 17: 763-71. mu-Slide Chemotaxis: a new chamber for long-term chemotaxis studies. 8. Behm B, Babilas P, Landthaler M, Schreml S. Cytokines, chemokines and BMC Cell Biol. 2011; 12: 21. growth factors in wound healing. J Eur Acad Dermatol Venereol. 2012; 26: 39. Gillitzer R, Goebeler M. Chemokines in cutaneous wound healing. J Leukoc 812-20. Biol. 2001; 69: 513-21. 9. Schreml S, Szeimies RM, Prantl L, Landthaler M, Babilas P. Wound healing in 40. Gallucci RM, Sloan DK, Heck JM, Murray AR, O'Dell SJ. Interleukin 6 the 21st century. J Am Acad Dermatol. 2010; 63: 866-81. indirectly induces keratinocyte migration. J Invest Dermatol. 2004; 122: 764-72. 10. Niethammer P, Grabher C, Look AT, Mitchison TJ. A tissue-scale gradient of 41. Boedtkjer E, Bunch L, Pedersen SF. Physiology, pharmacology and hydrogen peroxide mediates rapid wound detection in zebrafish. Nature. pathophysiology of the pH regulatory transport proteins NHE1 and NBCn1: 2009; 459: 996-9. similarities, differences, and implications for cancer therapy. Curr Pharm Des. 11. Stefonek TJ, Masters KS. Immobilized gradients of epidermal growth factor 2012; 18: 1345-71. promote accelerated and directed keratinocyte migration. Wound Repair 42. Pedersen SF, O'Donnell ME, Anderson SE, Cala PM. Physiology and Regen. 2007; 15: 847-55. pathophysiology of Na+/H+ exchange and Na+ -K+ -2Cl- cotransport in 12. Wenczak BA, Lynch JB, Nanney LB. Epidermal growth factor receptor the heart, brain, and blood. Am J Physiol Regul Integr Comp Physiol. 2006; distribution in burn wounds. Implications for growth factor-mediated repair. J 291: R1-25. Clin Invest. 1992; 90: 2392-401. 43. Cardone RA, Casavola V, Reshkin SJ. The role of disturbed pH dynamics and 13. Hinman CD, Maibach H. Effect of Air Exposure and Occlusion on the Na+/H+ exchanger in metastasis. Nat Rev Cancer. 2005; 5: 786-95. Experimental Human Skin Wounds. Nature. 1963; 200: 377-8. 44. Webb BA, Chimenti M, Jacobson MP, Barber DL. Dysregulated pH: a perfect 14. Winter GD. Formation of the scab and the rate of epithelization of superficial storm for cancer progression. Nat Rev Cancer. 2011; 11: 671-7. wounds in the skin of the young domestic pig. Nature. 1962; 193: 293-4. 45. Endres RG, Wingreen NS. Accuracy of direct gradient sensing by single cells. 15. Winter GD. Effect of Air Exposure and Occlusion on Experimental Human Proc Natl Acad Sci U S A. 2008; 105: 15749-54. Skin Wounds. Nature. 1963; 200: 378-9. 46. Firtel RA, Chung CY. The molecular genetics of chemotaxis: sensing and 16. Winter GD, Scales JT. Effect of air drying and dressings on the surface of a responding to chemoattractant gradients. Bioessays. 2000; 22: 603-15. wound. Nature. 1963; 197: 91-2. 47. Franca-Koh J, Devreotes PN. Moving forward: mechanisms of chemoattractant 17. Muller P, Schier AF. Extracellular movement of signaling molecules. Dev Cell. gradient sensing. Physiology (Bethesda). 2004; 19: 300-8. 2011; 21: 145-58. 48. Schneider IC, Haugh JM. Quantitative elucidation of a distinct spatial 18. Schreml S, Szeimies RM, Karrer S, Heinlin J, Landthaler M, Babilas P. The gradient-sensing mechanism in fibroblasts. J Cell Biol. 2005; 171: 883-92. impact of the pH value on skin integrity and cutaneous wound healing. J 49. Schneider IC, Haugh JM. Mechanisms of gradient sensing and chemotaxis: Eur Acad Dermatol Venereol. 2010; 24: 373-8. conserved pathways, diverse regulation. Cell Cycle. 2006; 5: 1130-4. 19. Horikoshi T, Balin AK, Carter DM. Effect of oxygen on the growth of human 50. Yang Y, Sourjik V. Opposite responses by different chemoreceptors set a epidermal keratinocytes. J Invest Dermatol. 1986; 86: 424-7. tunable preference point in Escherichia coli pH taxis. Mol Microbiol. 2012; 86: 20. Pentland AP, Marcelo CL. Modulation of proliferation in epidermal 1482-9. keratinocyte cultures by lowered oxygen tension. Exp Cell Res. 1983; 145: 51. Paradise RK, Lauffenburger DA, Van Vliet KJ. Acidic extracellular pH 31-43. promotes activation of integrin alpha(v)beta(3). PLoS One. 2011; 6: e15746. 21. Xia YP, Zhao Y, Tyrone JW, Chen A, Mustoe TA. Differential activation of 52. Paradise RK, Whitfield MJ, Lauffenburger DA, Van Vliet KJ. Directional cell migration by hypoxia in keratinocytes isolated from donors of increasing migration in an extracellular pH gradient: a model study with an engineered age: implication for chronic wounds in the elderly. J Invest Dermatol. 2001; cell line and primary microvascular endothelial cells. Exp Cell Res. 2013; 116: 50-6. 319: 487-97. 22. Schreml S, Szeimies RM, Prantl L, Karrer S, Landthaler M, Babilas P. Oxygen 53. Trompezinski S, Denis A, Vinche A, Schmitt D, Viac J. IL-4 and in acute and chronic wound healing. Br J Dermatol. 2010; 163: 257-68. interferon-gamma differentially modulate vascular endothelial growth factor 23. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg release from normal human keratinocytes and fibroblasts. Exp Dermatol. effect: the metabolic requirements of cell proliferation. Science. 2009; 324: 2002; 11: 224-31. 1029-33. 54. Sarangarajan R, Shumaker H, Soleimani M, Le Poole C, Boissy RE. Molecular 24. Walpole GS. An Improved Hydrogen Electrode. Biochem J. 1914; 8: 131-3. and functional characterization of sodium--hydrogen exchanger in skin as 25. Lee D, Khaja S, Velasquez-Castano JC, Dasari M, Sun C, Petros J, et al. In vivo well as cultured keratinocytes and melanocytes. Biochim Biophys Acta. 2001; imaging of hydrogen peroxide with chemiluminescent nanoparticles. Nat 1511: 181-92. Mater. 2007; 6: 765-9. 55. Halestrap AP. Monocarboxylic Acid transport. Compr Physiol. 2013; 3: 26. Nakai J, Ohkura M, Imoto K. A high signal-to-noise Ca(2+) probe composed of 1611-43. a single green fluorescent protein. Nat Biotechnol. 2001; 19: 137-41. 56. Wu L, Xia YP, Roth SI, Gruskin E, Mustoe TA. Transforming growth 27. Saito K, Chang YF, Horikawa K, Hatsugai N, Higuchi Y, Hashida M, et al. factor-beta1 fails to stimulate wound healing and impairs its signal Luminescent proteins for high-speed single-cell and whole-body imaging. Nat transduction in an aged ischemic ulcer model: importance of oxygen and age. Commun. 2012; 3: 1262. Am J Pathol. 1999; 154: 301-9. 28. Meier RJ, Schreml S, Wang XD, Landthaler M, Babilas P, Wolfbeis OS. 57. Sundaram GM, Common JE, Gopal FE, Srikanta S, Lakshman K, Lunny DP, et Simultaneous photographing of oxygen and pH in vivo using sensor films. al. 'See-saw' expression of microRNA-198 and FSTL1 from a single transcript Angew Chem Int Ed Engl. 2011; 50: 10893-6. in wound healing. Nature. 2013; 495: 103-6. 29. Schreml S, Meier RJ, Weiss KT, Cattani J, Flittner D, Gehmert S, et al. A 58. Burdick JA, Murphy WL. Moving from static to dynamic complexity in sprayable luminescent pH sensor and its use for wound imaging in vivo. Exp hydrogel design. Nat Commun. 2012; 3: 1269. Dermatol. 2012; 21: 951-3. 59. Neri D, Supuran CT. Interfering with pH regulation in tumours as a therapeutic strategy. Nat Rev Drug Discov. 2011; 10: 767-77.

http://www.thno.org Theranostics 2014, Vol. 4, Issue 7 735

60. Ludwig MG, Vanek M, Guerini D, Gasser JA, Jones CE, Junker U, et al. Proton-sensing G-protein-coupled receptors. Nature. 2003; 425: 93-8. 61. Tomura H, Mogi C, Sato K, Okajima F. Proton-sensing and lysolipid-sensitive G- protein-coupled receptors: a novel type of multi-functional receptors. Cell Signal. 2005; 17: 1466-76. 62. Arwert EN, Hoste E, Watt FM. Epithelial stem cells, wound healing and cancer. Nat Rev Cancer. 2012; 12: 170-80. 63. Harris KL, Bainbridge NJ, Jordan NR, Sharpe JR. The effect of topical analgesics on ex vivo skin growth and human keratinocyte and fibroblast behavior. Wound Repair Regen. 2009; 17: 340-6. 64. Kuchler S, Henkes D, Eckl KM, Ackermann K, Plendl J, Korting HC, et al. Hallmarks of atopic skin mimicked in vitro by means of a skin disease model based on FLG knock-down. Altern Lab Anim. 2011; 39: 471-80.

http://www.thno.org